From 7a5a9824c18f93415944c997dc6bb8eecfddd2e7 Mon Sep 17 00:00:00 2001 From: Denys Vlasenko Date: Wed, 3 Jun 2015 15:58:50 +0200 Subject: [PATCH] x86/asm/entry/32: Remove unnecessary optimization in stub32_clone Really swap arguments #4 and #5 in stub32_clone instead of "optimizing" it into a move. Yes, tls_val is currently unused. Yes, on some CPUs XCHG is a little bit more expensive than MOV. But a cycle or two on an expensive syscall like clone() is way below noise floor, and this optimization is simply not worth the obfuscation of logic. [ There's also ongoing work on the clone() ABI by Josh Triplett that will depend on this change later on. ] Signed-off-by: Denys Vlasenko Cc: Alexei Starovoitov Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Frederic Weisbecker Cc: H. Peter Anvin Cc: Josh Triplett Cc: Kees Cook Cc: Linus Torvalds Cc: Oleg Nesterov Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Thomas Gleixner Cc: Will Drewry Link: http://lkml.kernel.org/r/1433339930-20880-2-git-send-email-dvlasenk@redhat.com Signed-off-by: Ingo Molnar --- arch/x86/entry/ia32entry.S | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git a/arch/x86/entry/ia32entry.S b/arch/x86/entry/ia32entry.S index d0c7b28d5670..9558dacf32b9 100644 --- a/arch/x86/entry/ia32entry.S +++ b/arch/x86/entry/ia32entry.S @@ -529,14 +529,13 @@ GLOBAL(\label) GLOBAL(stub32_clone) leaq sys_clone(%rip), %rax /* - * 32-bit clone API is clone(..., int tls_val, int *child_tidptr). - * 64-bit clone API is clone(..., int *child_tidptr, int tls_val). - * Native 64-bit kernel's sys_clone() implements the latter. - * We need to swap args here. But since tls_val is in fact ignored - * by sys_clone(), we can get away with an assignment - * (arg4 = arg5) instead of a full swap: + * The 32-bit clone ABI is: clone(..., int tls_val, int *child_tidptr). + * The 64-bit clone ABI is: clone(..., int *child_tidptr, int tls_val). + * + * The native 64-bit kernel's sys_clone() implements the latter, + * so we need to swap arguments here before calling it: */ - mov %r8, %rcx + xchg %r8, %rcx jmp ia32_ptregs_common ALIGN -- 2.45.2